Ferric chloride assisted plasma pretreatment of

Bioresource Technology 243 (2017) 327–334

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Ferric chloride assisted plasma pretreatment of lignocellulose Rajeev Ravindran a, Chaitanya Sarangapani a, Swarna Jaiswal b, P.J. Cullen c, Amit K. Jaiswal a,⇑ a

School of Food Science and Environmental Health, College of Sciences and Health, Dublin Institute of Technology, Cathal Brugha Street, Dublin 1, Ireland Centre for Research in Engineering and Surface Technology, FOCAS Institute, Dublin Institute of Technology, Kevin Street, Dublin 8, Ireland C Department of Chemical and Environmental Engineering, University of Nottingham, Nottingham NG7 2RD, United Kingdom b

h i g h l i g h t s A novel FeCl3 assisted plasma pretreatment strategy was proposed for lignocellulose. Extensive delignification in spent coffee waste was achieved upon pretreatment. The polysaccharide fraction of spent coffee grounds was left unaffected. High reducing sugar yield was obtained upon hydrolysis of pretreated spent coffee waste. Pretreated SCW was found to be a suitable substrate for bioethanol production.

a r t i c l e

i n f o

Article history: Received 25 April 2017 Received in revised form 21 June 2017 Accepted 22 June 2017 Available online 26 June 2017 Keywords: Lignocellulose Biomass pretreatment Spent coffee waste Cold air pressure plasma Fenton reaction Bioethanol production

a b s t r a c t In this study, a novel pretreatment for spent coffee waste (SCW) has been proposed which combines two techniques viz. atmospheric air plasma and FeCl3 to create a superior pretreatment that involves Fenton chemistry. The pretreatment was optimised employing Taguchi Design of Experiments, and five parameters were taken into consideration viz. biomass loading, FeCl3 concentration, H2SO4 concentration, plasma discharge voltage and treatment time. The composition analysis of the pretreated SCW revealed substantial amounts of lignin removal, with a maximum for process conditions of 70 kV for 2 min in an acidic environment containing 1% H2SO4. FTIR, XRD and DSC were performed to characterise the samples. The pretreated SCW after enzymatic hydrolysis yielded 0.496 g of reducing sugar/g of SCW. The hydrolysate was subjected to fermentation by S. cerevisiae and led to the production of 18.642 g/l of ethanol with a fermentation efficiency of 74%, which was a two fold increase in yield compared to the control. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction With petroleum resources dwindling, the world has a renewed interest in sustainable sources of energy such as bioethanol. Lignocellulose is one the most abundant renewable resources in the world. In recent years, lignocellulosic plant wastes from various industry sectors such as food and agriculture have widely been researched as potential substrates for bioethanol production. Spent coffee waste (SCW) is the solid material obtained after brewing of instant coffee and is a rich source of lignocellulose. Over the past few years’ global coffee production and consumption has been on the rise. Coffee is one of the most popular beverages in the world, with global coffee production reaching 8.5 billion tonnes in 2015/2016 (ICO, 2016). Lignocellulose is a complex polymer of polysaccharides and lignin. In nature, plant cell wall can naturally resist degradation by ⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (A.K. Jaiswal). http://dx.doi.org/10.1016/j.biortech.2017.06.123 0960-8524/Ó 2017 Elsevier Ltd. All rights reserved.

microbial, enzymatic and chemical attack. Consequently, the efficient utilisation of polysaccharides is challenging. This calls for suitable methods that can remove recalcitrance of lignocellulose. Pretreatment of lignocellulosic biomass are necessary to disrupt the lignocellulosic complex to expose cellulose to enzymatic hydrolysis (Ravindran and Jaiswal, 2016a). Numerous studies have reported on pretreatment methods and their efficacies over the past few decades. Some of the most effective pretreatment methods that offer commercial feasibility include steam explosion, dilute acid pretreatment and ammonia fibre expansion (Campbell et al., 2013; Rocha et al., 2015). Recently, treatment with ferric chloride was investigated as a feasible pretreatment measure for bagasse, rice straw and wood fibre by Chen et al. (2015). Studies involving plasma generated ozone for delignification of wheat straw and subsequent bioethanol production was studied by Schultz-Jensen et al. (2011). Several pretreatment methods have been devised for lignocellulose over the past decade. However, very few studies have focused the effective pretreatment of SCW. For example, characterisation of

328

R. Ravindran et al. / Bioresource Technology 243 (2017) 327–334

polysaccharides extracted from alkali pretreated SCW was performed by Ballesteros et al. (2015). Conde and Mussatto (2016) studied the effectiveness of hydrothermal pretreatment strategy in the isolation of polyphenols from SCW. Similarly, Scully et al. (2016), applied hydrothermal pretreatment for the industrially important sugars. However, the authors were unable to come across any study that involved mimicking Fenton reaction using plasma for lignocellulosic waste such as spent coffee waste and its potential application in bioethanol production. Plasma is the fourth state of matter. It contains atoms, ions and molecules in a metastable state with a net electric charge of zero (Devi et al., 2017). Atmospheric plasma discharges may be obtained artificially by many means of electromagnetic wave disturbances resulting from the application of current or ionisation radiation in way such that a measurable charge is created in the gas without the generation of any heat. This gives rise to positively and negatively charged electrons, intermediate highly reactive species (H, O, OH etc.) atoms, molecules and UV photons with neutral charge (Sarangapani et al., 2015; Sarangapani et al., 2016a,b). Ozone, one of the reactive species generated because of atmospheric air plasma specifically degrades lignin when lignocellulosic materials are subjected to plasma treatment. This leaves the solid fraction rich in polysaccharides which can be utilised. Studies involving plasma as a pretreatment measure have reported high dry matter concentration, relatively low pretreatment time, low temperature of operation and absence of any inhibitory compounds as the major advantages. The phenolic compounds that arise from the lignin degradation can be removed by simply washing the solid material with water (Souza-Corrêa et al., 2014). Fenton chemistry is an oxidation reaction which gives rise to highly reactive free radicals. In the presence of a Fe3+ ion and H2O2 the ferric ion is oxidised to an Fe2+ ion resulting in the formation of H2O and O2 along with two reactive species viz. HO and HOO (Fenton, 1894; Pignatello et al., 2006). The Fenton reaction does not require high temperatures, pressure or any other chemical in the form of a catalyst for initiation. However, it does require an acidic environment (Jung et al., 2009). The generation of reactive species in the presence of Fe3+ can lead to the occurrence of Fenton chemistry which can be used for the depolymerisation of lignin in plant biomass. Several studies have reported where the Fenton reaction has been exploited as a pretreatment measure (Jung et al., 2015; Kato et al., 2014). Pretreatments can involve several parameters to achieve for effective delignification of lignocellulose. In most common pretreatments such as dilute acid hydrolysis or organosolv pretreatments, these parameters include acid concentration, temperature, pressure, ethanol concentration etc. (Ravindran et al., 2017a). The abundance of parameters makes it difficult to determine the optimal levels of each factor to attain the target objective. Taguchi design of experiments describes an effective method to evaluate the effects of different parameters statistically in a process that has multiple factors. Furthermore, it minimises the errors caused due to variations in the experimental process. Taguchi design of experiments works by testing the effects of varying parameters simultaneously. This allows the accurate estimation of individual factors thus selecting the best combination for factors that will result in a robust process (Azad et al., 2016). In the present study, we investigate a new pretreatment strategy for lignocellulose by combining a nonthermal plasma discharge and Fe3+ ions to create an environment for the Fenton reaction along with ozone to delignify lignocellulose, and spent coffee waste was chosen as a model. The pretreatment process was optimised by the Taguchi Design of Experiments that involved five parameters viz. biomass loading, acid concentration, ferric chloride concentration, voltage and time. Furthermore, the pretreated SCW was enzymatically hydrolysed using the optimised

parameters and the hydrolysate obtained was employed for bioethanol production.

2. Materials and methods 2.1. Feedstock and other chemicals Spent Coffee Waste (SCW) was collected from a local coffee outlet located in Dublin city. The waste material was dried in a hot air oven at 80 °C for 48 h (Kwon et al., 2013). The dried SCW was then stored at room temperature in a cool and dry place for further experiments. All the chemicals such as FeCl3, sulphuric acid and saccharifying enzymes such cellulase from Trichoderma reesei and hemicellulase from Aspergillus niger were purchased from Sigma Aldrich, Ireland. Cellulase was purchased in liquid form. Following protocols devised by National Renewable Energy Laboratory (Adney and Baker, 1996), the cellulase activity was assayed. The hemicellulase, on the other hand, was obtained as powder form and dissolved in sodium acetate buffer (pH 4.8, 50 mM) to make up a concentration of 10 g/l. The subsequent activity of the enzyme was assayed followed protocols described by Rickard and Laughlin (1980). Cellulase enzyme registered an enzyme activity of 77 FPU/ ml while hemicellulase showed 72 U/ml enzyme activity. 2.2. Component analysis The compositional variations in SCW before and after pretreatment was analysed by using protocols devised by the National Renewable Energy Laboratory, USA. As a control measure the composition of native SCW was also determined. Biomass was hydrolysed by mixing it with 72% H2SO4 for 60 min at 30 °C. The acid concentration was then diluted to 4% with deionised water and autoclaved at 121 °C for 1 h. The solids were then separated from the liquids and dried at 80 °C for 48 h following which the acid insoluble lignin was determined by drying at 595 °C for 24 h. The liquid was assessed for the acid soluble lignin by measuring the absorbance at 205 nm and the sugar concentrations were determined by HPLC (Rezex ROA H+, Waters e2695 Separation module, Refractive Index detector). 2.3. Surface modification using atmospheric air pressure plasma Various parameters affecting the metal chloride assisted plasma pretreatment were optimised by adopting a Taguchi design. Five parameters viz. biomass loading, ferric chloride concentration, dilute acid concentration, time and voltage were taken into consideration for experimental design. Three levels were considered for each parameter. Based on these parameters and levels an L27 orthogonal array of standard configuration of five 3 level factors was set up (Table 1) and was generated by using Minitab 17Ò. A prototype high voltage but low power (150 W) dielectric barrier discharge plasma source was used for this study. The system consisted of a high voltage transformer (230 V, 50 Hz), a ground electrode and voltage variac (output voltage can be controlled from 0 to 120 kV). The optimisation trials were operated at 60, 70 and 80 kV as per the experiment design. The input voltage and current was monitored by employing an oscilloscope (InfiniVision 2000 XSeries Oscilloscope, Agilent Technologies Inc., USA) (Fig. 1). The samples were held in the centre of a polypropylene container by means of a petri dish. On completion of each trail the containers were sealed in polypropylene bags (B2630; Cryovac Sealed Air Ltd, Dunkan, SC, USA) and stored in a cool and dry place for the reaction to take place (Lu et al., 2014). After pretreatment the solids were filtered out. Any residual FeCl3 was removed by wash-

329

R. Ravindran et al. / Bioresource Technology 243 (2017) 327–334 Table 1 Taguchi design for optimisation of process parameters involved in plasma pretreatment. Trial No.

Biomass loading (%w/v)

FeCl3 conc.%

H2SO4 conc.%

Time (min)

Voltage (kV)

Reducing sugar (g/g)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 10 10 10 10 10 10 10 10 10 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5

1 1 1 2 2 2 3 3 3 1 1 1 2 2 2 3 3 3 1 1 1 2 2 2 3 3 3

1 1 1 1.4 1.4 1.4 1.6 1.6 1.6 1.4 1.4 1.4 1.6 1.6 1.6 1 1 1 1.6 1.6 1.6 1 1 1 1.4 1.4 1.4

2 2 2 4 4 4 6 6 6 6 6 6 2 2 2 4 4 4 4 4 4 6 6 6 2 2 2

60 70 80 60 70 80 60 70 80 60 70 80 60 70 80 60 70 80 60 70 80 60 70 80 60 70 80

0.299 0.311 0.317 0.296 0.302 0.311 0.300 0.313 0.318 0.253 0.264 0.272 0.309 0.309 0.321 0.413 0.428 0.455 0.269 0.296 0.300 0.428 0.452 0.446 0.432 0.493 0.474

Fig. 1. Instrument set up for the generation of atmospheric plasma.

ing the solids five times with deionised water. They were then freeze dried and stored for further experiments.

supernatant was collected and were checked for reducing sugar content using dinitrosalicylic acid and the values obtained were input into the design software to decipher optimised hydrolysis conditions.

2.4. Enzymatic hydrolysis The parameters for maximum reducing sugar production was derived from a study performed previously in our laboratory (Ravindran et al., 2017b). Briefly, the pretreated SCW was hydrolysed by employing a biomass loading of 1%, cellulase and hemicellulase loading of 1.5 ml (23.1 FPU/g of dry biomass) and 0.37 ml (5.33 U/g of dry biomass), respectively while maintaining a pH of 6.7. The ambient temperature for the reaction was maintained at 50 °C while the reaction volume was set to 50 ml. On completion of the experiments the suspensions were transferred to 30 ml polypropylene tubes, centrifuged at 7000 rpm for 10 min. The

2.5. Individual sugar, inhibitor and organic acid analysis The presence and quantification of monosaccharides and organic acids was performed using an Alliance HPLC (Waters, e2695 Separation module) with a Rezex ROA-Organic acid H+(8%) column, (350 7.8 mm; Phenomenex, UK) and 0.005 M H2SO4 as the mobile phase at 65 °C maintaining a flow rate of 0.6 ml/min (Jaiswal et al., 2012). The same type of guard column was used with the regular column and was kept outside the compartment to avoid overheating beyond the manufacturers recommended

330


limit (60 °C). An isocratic mobile phase of 0.01 M sulphuric acid was used to detect and estimate the number of inhibitors such as furfural and hydroxymethyl furfural in the hydrolysate. The HPLC system was equipped with an autosampler, degasser and a UV detector (210 nm) and a refractive index (RI) detector. 2.6. Bioethanol production The sugar rich hydrolysate, free from undissolved solids, obtained after saccharification was subjected to fermentation to produce bioethanol. Saccharomyces cerevisiae obtained from the DIT microbiology repository was employed as the fermentative microorganism. For inoculation, the microbe was grown in a medium containing glucose (1 g/100 ml), and yeast extract (0.1 g/100 ml) supplemented by (NH4)2SO4 (0.5 g/100 ml), MgSO47H2O (0.05 g/100 ml) and KH2PO4 (0.1 g/100 ml) and was incubated at 30 °C and 120 rpm for 18 h. For bioethanol production, 0.5 ml of actively growing culture of S. cerevisiae was added into the reaction mixture. The reaction mixture consisted of the hydrolysate which was supplemented with yeast extract (0.1%) and peptone (0.1%) was autoclaved at 121 °C for 15 min. The pH was maintained by the addition of 1% (v/v) acetate buffer (pH 5.4, 20 mM). Fermentation was carried out in 100 ml conical flasks with a total reaction volume of 50 ml at 30 °C for 72 h. After fermentation, the reaction mixture was centrifuged and filtered using 0.4 l filters and the ethanol produced was quantified by using Gas Chromatography [Bruker Scion 456-GC coupled with flame ionisation detector (FID)]. The injector temperature was set at 220 °C while the column oven was set at an initial temperature of 80 °C. which was ramped up to 160 °C at a rate of 40 °C per min and held for 7 min. A flame ionisation detector was used to analyse the ethanol content and set at a temperature of 200 °C. 2.7. Characterisation of pretreated SCW 2.7.1. X-ray diffraction X-ray diffraction studies were employed to study any changes imparted in the crystallinity of the lignocellulose substrate by the pretreatment strategy (Ravindran et al., 2017c) XRD of the pretreated and native SCW was performed in a Siemens D-500 X-ray diffractometer. The voltage and current for the radiation were set at 40 kV and 30 mA using Cu Ka as the radiation source (k = 0.154 nm). The diffraction angles were spanned at 2h = 5°–50°. 2.7.2. FTIR analysis FTIR spectroscopy was performed on raw and pretreated SCW to identify possible structural changes as a reflection of the variations in the functional groups in the SCW before and after pretreatment. A Perkin Elmer Spectrum GX FT-IR (UATR) Microscope (USA) was employed in this study as per the method described in Ravindran et al. (2017c). In brief, the FTIR spectra for SCW samples were recorded from 4000 to 400 cm1 with 16 scans at a resolution of 0.3 cm1 in transmission mode. 2.7.3. Thermal behaviour study using DSC Differences in the thermal behaviour of SCW after pretreatment was studied using differential scanning calorimetry (DSC), using 55 mg of the pretreated SCW in an aluminium pan with an empty pan used as a reference. The thermograms were obtained by increasing the temperatures from 20 °C to 500 °C at a rate of 10 °C per min at a constant nitrogen atmosphere. All measurements were carried out between 25 °C and 500 °C with a linear increase of 10 °C in a Shimadzu DSC-60 installed with TA-60WS software.

2.8. Estimation of total phenolic content The total phenolic content in the pretreatment liquor was determined by following protocol described by (Jaiswal and AbuGhannam, 2013). Briefly, 100 ll of sample was mixed with 2 ml of 2 mM Na2CO3 solution and incubated for 2 min at room temperature. 100 ll of 1 N Folin-Ciocalteau’s phenol reagent was added followed by incubation for 30 min in darkness at room temperature. After the reaction was complete, the absorbance was measured at 720 nm. The results were expressed as microgram per ml of the sample through calibration curve of gallic acid. 2.9. Determination of hydrogen peroxide concentration Hydrogen peroxide formed because of plasma treatment was quantified by oxidising potassium iodide measuring the amount of iodine formed using spectrophotometry at 390 nm. In brief, 100 ll of 1 M KI solution and 50 ll of phosphate buffer were mixed with 50 ll of plasma treatment liquor. The mixture was incubated for 30 min after which the absorbance was measured at 390 nm (Boehm et al., 2016). 2.10. Optical emission spectroscopy Optical emission spectroscopy was performed to determine the active chemical species formed during the atmospheric air plasma discharge. This was carried out using a Stellarnet EPP 2000C-25 spectrometer at a resolution of 1.5 nm. The light from the plasma was coupled via an optical fibre. The diffraction grating in the spectrometer had a radius of curvature of 40 mm with 590 grooves per mm and an entrance slit width of 25 mm. The fibre (numerical aperture = 0.22) was optimised for performance in the ultraviolet and visible region of electromagnetic spectrum. The spectrometer operated within the wavelength range of 190 nm–850 nm. The integration time was set at 5000 ms and 5 samples were averaged for the collection of spectra. The emission spectra were qualitatively analysed to assign chemical species to the peaks. The noise in the spectra was cancelled, averaged and analysed using National Institute of Standards and Technology (2013) atomic spectra database and literature published elsewhere (Meiners, 2010). 2.10.1. Statistical analysis Quantitative experimental tests were carried out in triplicate and the results are reported as the mean ± the standard deviation. The Taguchi statistical analysis was conducted by employing MINITAB 17 software. To evaluate the sensitivity of each parameter, determine the optimum combination of operating parameters a variance analysis (ANOVA) was performed at confident level of 95%. 3. Results and discussion 3.1. Effect of pretreatment on the composition of spent coffee waste The Fenton reaction involves the generation of hydroxyl radicals or similarly powerful oxidising agents in the presence of a metal ion such as iron or hydrogen peroxide. One of the several chemical reactions involved in Fenton chemistry includes the depolymerisation of lignin by dealkylation which then facilitates the enzymatic digestion of cellulose and hemicellulose components (Arantes et al., 2012). This novel pretreatment strategy attempts to delignify SCW by means of providing a suitable environment for the Fenton reaction to take place by providing ferric ions and hydrogen peroxide while maintaining a low pH. Composi-


tion analysis of native and plasma pretreated SCW was performed to determine the changes in individual components brought about by the pretreatment strategy. Native SCW was rich in glucomannan and galactomannan content which was evident from the galactose (13.7 ± 0.3 g/100 g of SCW) and mannose (21.2 ± 0.5 g/100 g of SCW) content with respect to glucose (8.6 ± 0.1 g/100 g of SCW). Trace amounts of arabinose was also found in SCW (1.7 ± 0.2 g/100 g of SCW). Similar findings were reported by (Ballesteros et al., 2015; Ballesteros et al., 2014). On the other hand, the plasma pretreated SCW consisted of 18.76 ± 0.2 g of galactose, 24.51 ± 1.3 g of mannose, 10.74 ± 0.9 g of glucose and 2.24 ± 1.2 g of arabinose per 100 g of SCW. From the results, it is evident that there was an increase in the total polysaccharide content in SCW after the pretreatment. As with lignin content, the acid insoluble lignin reduced from 31.12 ± 0.2 g in native SCW to 18.6 ± 0.45 g in pretreated SCW (per 100 g SCW).

3.2. Effect of different process parameters on metal chloride assisted plasma pretreatment Table 1 represents the Taguchi experimental design and associated reducing sugar results obtained in the study. Results showed a maximum reducing sugar yield was obtained from trial No. 26 (0.493 g/g of SCW) (refer Table 1). The pretreatment parameters for this trial was found to be 12.5% biomass loading, 3% FeCl3 concentration, 1% H2 SO4 concentration, exposure time of 2 min and a voltage of 70 kV. Fig. 2 comprises of contour plots that depict the interactions of the process parameters considered in this study that contribute to the release of reducing sugar after enzymatic hydrolysis of the pretreated SCW. A general trend observed was that a higher biomass content resulted in higher reducing sugar concentration after enzymatic hydrolysis. From Fig. 2a, it is evident that a high biomass content (% w/v) exposed to plasma for lower time periods (min) was favourable in enhancing the reducing sugar yield. At a biomass concentration of 12.5% (w/v) and a plasma exposure time of around 2 min the reducing sugar release was found to be in the range of 0.40 g/g of pretreated SCW. Increasing the exposure time detrimentally effected the reducing sugar yield. This was possibly because of the extensive damage incurred by the

331

polysaccharide fraction along with lignin which left less cellulose and hemicellulose for enzymatic degradation. Fig. 2b represents the effects of voltage and biomass on the eventual release of the reducing sugar from SCW in the form on a contour plot. From the plot, it is evident that low voltages and low biomass loading rates (8–10%) was largely ineffective in breaking down the complex lignocellulose structure as the reducing sugar yield was low. As the voltage increases (70–80 kV) the effectiveness of the pretreatment increases. At higher voltages, higher concentrations of reactive species are generated which in turn accelerates the delignifying degradation process (Moiseev et al., 2014). Fig. 2c shows the relationship between H2SO4 concentration and biomass. A lower concentration of acid (1% acid conc.) was found to favour higher reducing sugar formation. This could be because the higher concentrations of acid tend to produce inhibitory compounds which negatively affect enzymatic hydrolysis. Fig. 2d shows the interaction between FeCl3 concentration and biomass loading. A higher concentration of FeCl3 resulted in higher reducing sugar yields. This may be because of the accelerated generation of reactive radicals are a result of the Fenton chemistry due to higher concentrations of Fe3+ ions.

3.3. Statistical analysis and regression model equation The ANOVA outputs performed on the L25 Taguchi Orthogonal Array is provided in Table 2. ANOVA studies gave insights on the role of each parameter based on Fischer’s test (F-value), probability and sum of squares to check the significance of each parameter in the model. A large F-value is suggestive of the importance of a parameter applied in the process. Accordingly, three parameters viz. biomass loading, FeCl3 conc. and H2SO4 conc. were found to have higher F-values compared to voltage and time. Acid concentration had the most prominent effect on the reducing sugar yield with an F-value of 244.84 followed by FeCl3 conc. (F-value = 151.2) and lastly biomass loading (F-value = 113.04) Furthermore, only three parameters (biomass loading, FeCl3 conc. and H2SO4 conc.) were found to be significant (p < 0.05) with respect to influencing reducing sugar yield. Voltage did not have a significant impact (p > 0.05) on the effectiveness of the pretreatment. A similar

Fig. 2. Contour plots exhibiting the interactions of various process parameters on the reducing sugar yield (g/g); (a) interactions between biomass (% w/v) and time (min); (b) interactions between biomass (% w/v) and voltage (kV); (c) interactions between biomass and H2SO4 concentration (%); (d) interactions between FeCl3 concentration (%) and biomass (% w/v).

332


Table 2 Analysis of variance data for reducing sugar yield after plasma pretreatment. Source

DF

Adj SS

Adj MS

F-Value

P-Value

Biomass loading FeCl3 H2SO4 Time Voltage Error Total

1 1 1 1 1 1 6

0.037545 0.060517 0.036434 0.002668 0.00254 0.005191 0.144895

0.003833 0.0061759 0.0037195 0.002723 0.002596 0.000687

113.04 151.2 244.84 147.41 89.95 54.83

0.019 0.006 0.014 0.007 0.06 0.065

R-sq = 96.42%

R-sq(adj) = 95.56%

S = 0.0157216 2

S-square of root mean square; R -coefficient of determination; Adj SS-adjusted sum of squares; Adj MS-adjusted mean square; F-F test; P-probability.

observation was reported by Misra et al. (2015), where atmospheric air pressure plasma was used to study the variations in anthocyanin content in strawberries. The variability in observations of this study can be attributed to the change in plasma chemistry due to variations in the nature of the lignocellulosic substrate which have not been included in this study. A regression model equation was generated as a part of the ANOVA study which comprised all the factors considered in this study. This equation can predict the reducing sugar yield according to any set of parametric settings within the range of the process parameters. The regression equation is as follows:

Reducing sugar ðg=gÞ ¼ 0:1864 þ 0:01827 biomass loading þ 0:05798 FeCl3 conc: 0:1473 H2 SO4 conc: 0:00609 time þ 0:001188 v oltage From a high R2 value of 96.42% and the adjusted R2 of 95.56% illustrated that the model adequately fit the data. 3.4. Incidence of inhibitors in pretreatment liquor The pretreatment liquor was subjected to HPLC analysis to determine any incidence of inhibitory or monosaccharide components. As was expected trace amounts of individual sugars such as glucose, galactose and mannose and, a small fraction of cellobiose was found in the liquor. An interesting finding was the presence of hydroxymethyl furfural (HMF) in the treatment liquors which were high in sulphuric acid concentration. The concentration ranges of HMF in pretreatment liquors ranged from 0.5 g to 1.2 g/100 g of SCW. This may have been detrimental in the enzymatic hydrolysis of the pretreated SCW. It has been well documented that glucose in acidified solutions can give rise to hydroxymethyl furfural (Van Dam et al., 1986). The inhibitory compounds may have been formed by the degradation of glucose by acid. The increasing intensity of plasma coupled with the higher exposure times may have accelerated the process of HMF formation. 3.5. FTIR, XRD and DSC profiles of untreated and pre-treated spent coffee waste Fourier Transform Infrared Spectroscopy is used to identify or analyse the changes induced by a pretreatment process on lignocellulosic residues. Each component in lignocellulose (cellulose, hemicellulose and lignin) will have bonds and functional groups that are specific to each of them. Any changes in structure such as the breakage of bonds or removal of fractions can result in changes that can identified in the FTIR spectra. The peaks at 897 cm1 represents glycosidic linkage between cellulose and hemicellose components. The peak intensity was considerably lower in the spectrum of the pretreated SCW. The absorbance of

band at 1035 cm1, which indicates CAO, C@C and CACAO stretching between polysaccharides and lignin decreased after pretreatment (Tamaki and Mazza, 2011). Furthermore, the decrease in band 1200 cm1 indicated the breakage of hydrogen bonds between cellulose and hemicellulose. The peak at 1290 cm1 is associated with the crystallinity of SCW (Binod et al., 2012). This peak was found to be diminished in the pretreated SCW spectrum. The band at 1400 cm1 is representative of the lignin present in the sample. This band was seen to be considerably reduced which indicates that the plasma pretreatment was effective in lignin removal. Band 1750 cm1 is an indication of the ketone/aldehyde bonds in hemicellulose (Bodirlau and Teaca, 2009). A peak for this region was found in the pretreated sample spectrum, albeit in lower intensity, probably due to the stretching of these bonds. The band 2920 cm1 represents methyl and methylene groups in lignocellulose (Haripriya et al., 2014). These bands were not as prominent in the pretreated sample spectra, which is an indication of possible demethylation of SCW by reactive radicals generated during the plasma treatment. The absence of peaks in the band range 3000– 3500 cm1 is indicative of stretching of –OH groups for the pretreated SCW (Jahan et al., 2011). Lignocellulose is generally crystalline as well as amorphous in nature due to its chemical composition. Cellulose is the component in plant biomass that provides its crystallinity. Crystalline cellulose is more resistant to degradation due to chemical, enzymatic or microbial attack compared to its amorphous counter-part. Conversely, hemicellulose is amorphous in nature and is susceptible to degradation by chemical agents or microorganisms. The efficient utilisation of cellulose involves a shift from its crystalline nature to amorphous form (Ravindran and Jaiswal, 2016b). X-ray diffraction studies help to understand the crystallinity of the lignocellulose. Peaks at 2h = 22° are indicative of the crystallinity of the material. Meanwhile, peaks at 2h = 18° represents the amorphous regions in a lignocellulosic substrate. Native SCW was found to be a mixture of equally crystalline and amorphous with broad peaks ranging from 2h = 15° to 2h = 25°. However, in the plasma pretreated SCW there was an absence of such peaks in the crystalline region. Coupled with the development of peaks in the amorphous region it is clear that the plasma treatment leads to decrease in crystallinity of lignocellulose. Although not many studies are available on the changes brought about by air plasma on lignocellulosic fibres, (Baltazar-Y-Jimenez and Bismarck, 2007) reported that the crystallinity of cellulose fibres is affected by the plasma treatment. Thermal behaviour studies of lignocellulose before and after pretreatment were conducted to analyse the changes in the biomass in terms of its properties as a polymer. This was performed using differential scanning calorimetry (DSC). DSC provides insights into the physical and chemical characteristics of lignocellulose. Several properties such as crystallisation, heat capacity, melting, crystalline orientation and glass transition can be


recorded using DSC. The changes induced by the pretreatment in the thermal behaviour of lignocellulose has been widely studied by several researchers (Ballesteros et al., 2014; Nguyen et al., 1981). Therefore, DSC was performed to study the changes induced in the physical properties of SCW after plasma pretreatment. The DSC thermograms are provided as Fig. 3. The thermogram for native SCW was marked by an exothermic event between 20 °C and 102.9 °C with an associated enthalpy of 114.99 J/g. This may have occurred because of vaporisation of water and the crystalline nature of the sample. The exothermic event was followed by a phase transformation inducing a change in the heat capacity. The glass transition temperature was recorded at 259.2 °C. The effect of plasma on the physical properties of SCW was evident as there were marked differences between the thermograms of the pretreated and native SCW. The DSC thermogram for plasma treated SCW showed distinct peaks that suggested an event for glass transition, crystallisation, melting and ending transient. The glass transition temperature was recorded at 283 °C.

3.6. Estimation of total phenolic content and hydrogen peroxide in pretreatment liquor Exposing aqueous solutions to plasma discharge results in the generation of large amounts of hydrogen peroxide. This results in the initiation of Fenton reaction that further degrades individual components in lignocellulose such as lignin. The hydrogen peroxide in the pretreatment liquor formed was quantified. It was found that the liquid obtained after pretreatment contained 647.28 lM of hydrogen peroxide after 24 h of plasma treatment. A majority of the H2O2 formed is due to its solubilisation from gaseous phase into liquid phase. Therefore, by retaining the reactive species in the gas phase in contact with the liquid phase for longer periods time higher amounts of H2O2 could diffuse into the liquid. Furthermore, earlier studies confirmed that concentrations of H2O2 in solutions subjected to plasma treatment were stable over several weeks when stored in closed containers at 4 °C (Boehm et al., 2016). Spent coffee waste is high in phenolic compounds. The effect of plasma on the phenolic content of pretreatment liquor was analysed by Folin-Ciocalteau’s phenol reagent. The control contained 908.35 lg GAE/g of SCW while the pretreated liquor contained 672.86 lg GAE/g of SCW. Exposing SCW to dielectric barrier discharge plasma resulted in the total phenolic content by 26% for an experimental setting of 70 kV and 2 min. The reduction in total phenolic content can be attributed to the formation of reactive species and the availability of oxygen in the atmosphere which actively degraded phenolic compounds in SCW (Ramazzina et al., 2016). Earlier studied conducted by Grzegorzewski et al. (2011)

333

reported a reduced phenolic content in lamb lettuce leaves when treated with atmospheric air pressure plasma. 3.7. Identification of radical species formed during atmospheric air plasma The chemical reactive species in the gas phase formed during the atmospheric air plasma was analysed by optical emission spectroscopy (OES). The spectrum of the radiation emitted by the plasma was grated and the intensity was measured as function of the wavelength (Fig. 4). The non-thermal plasma was operated at voltage of 70 kV the emission spectra was obtained for a wavelength range of 180–900 nm. The emission spectrum revealed that the emission was near the UV region (300–400 nm). Emissions from N2 and N+2 species exhibited distinct peaks. The small peaks observed at the region corresponding to 250–300 nm were a resultant of the presence of OH. Meanwhile the very low intensity peaks observed at 750 nm and 780 nm corresponded to singlet oxygen. The formation of singlet O may be attributed to the particle collisions as well as the quenching of O(3P) and O(5P) energy. Similar observations were reported by Pearse and Gaydon (1976) and Laux et al. (2003) in different studies involving identification of molecular spectra and atmospheric air pressure plasma. From these inferences, it was quite clear that a certain amount of lignin degradation was achieved by reactive species formed during plasma treatment. 3.8. Bioethanol production using plasma treated SCW SCW was found to be a potential candidate for bioethanol production. The hydrolysate obtained after enzymatic digestion of the plasma pretreated SCW was used for bioethanol production using S. cerevisiae. A maximum ethanol content of 18.642 g/l was obtained after fermentation for 72 h. The bioethanol production using the control sample yielded 9.231 g/l of ethanol. A 2-fold increase in bioethanol yield using the optimised process parameters was found with a fermentation efficiency of 74%. This result is inline with previous studies carried out by Kwon et al. (2013), which reported a high ethanol fermentation efficiency of 80% using lipid extracted SCW. Similar findings were reported by Sindhu et al. (2011) where they optimised dilute acid pretreatment for the processing of sugar cane trash and eventual bioethanol production to obtain a titre of 11.365 g/l.

70 60 50

mW

40 30 20 10 0 -10 -20

0

100

200

300

400

500

Temperature (°C) Naive SCW

Pretreated SCW

Fig. 3. DSC thermogram of native and plasma pretreated SCW.

Fig. 4. Typical Optical Emission Spectrum (OES) of the dielectric barrier discharge in air (Operating voltage = 70 kV).

334


4. Conclusion Ferric chloride assisted plasma processing of SCW was found to be an effective pretreatment strategy with high lignin removal capacity. This resulted in an enhanced enzymatic digestion and high fermentation efficiency for ethanol production. A Taguchi design of experiments was successful in optimising the parameters for the plasma treatment. A maximum effectiveness of the pretreatment was achieved after a short treatment duration of 2 min at a voltage of 70 kV and in the presence of 1% H2SO4. To our knowledge there are limited experiments employing atmospheric air plasma as a lignocellulose pretreatment measure let alone trying to mimic Fenton chemistry using plasma for delignification of plant biomass and subsequent bioethanol production. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2017.06. 123. References Adney, B., Baker, J., 1996. Measurement of cellulase activities. LAP. 6. Arantes, V., Jellison, J., Goodell, B., 2012. Peculiarities of brown-rot fungi and biochemical Fenton reaction with regard to their potential as a model for bioprocessing biomass. Appl. Microbial. Biotechnol. 94, 323–338. Azad, F.N., Ghaedi, M., Dashtian, K., Hajati, S., Pezeshkpour, V., 2016. Ultrasonically assisted hydrothermal synthesis of activated carbon–HKUST-1-MOF hybrid for efficient simultaneous ultrasound-assisted removal of ternary organic dyes and antibacterial investigation: Taguchi optimization. Ultrason. Sonochem. 31, 383– 393. Ballesteros, L.F., Teixeira, J.A., Mussatto, S.I., 2014. Chemical, functional, and structural properties of spent coffee grounds and coffee silverskin. Food Bioprocess. Technol. 7, 3493–3503. Ballesteros, L.F., Cerqueira, M.A., Teixeira, J.A., Mussatto, S.I., 2015. Characterization of polysaccharides extracted from spent coffee grounds by alkali pretreatment. Carbohydr. Polym. 127, 347–354. Baltazar-Y-Jimenez, A., Bismarck, A., 2007. Surface modification of lignocellulosic fibres in atmospheric air pressure plasma. Green Chem. 9, 1057–1066. Binod, P., Satyanagalakshmi, K., Sindhu, R., Janu, K.U., Sukumaran, R.K., Pandey, A., 2012. Short duration microwave assisted pretreatment enhances the enzymatic saccharification and fermentable sugar yield from sugarcane bagasse. Renewable Energy 37, 109–116. Bodirlau, R., Teaca, C.A., 2009. Fourier transform infrared spectroscopy and thermal analysis of lignocellulose fillers treated with organic anhydrides. Rom. J. Phys. 54, 93–104. Boehm, D., Heslin, C., Cullen, P.J., Bourke, P., 2016. Cytotoxic and mutagenic potential of solutions exposed to cold atmospheric plasma. Sci. Rep. 6, 21464. Campbell, T.J., Teymouri, F., Bals, B., Glassbrook, J., Nielson, C.D., Videto, J., 2013. A packed bed ammonia fiber expansion reactor system for pretreatment of agricultural residues at regional depots. Biofuels 4, 23–34. Chen, L., Chen, R., Fu, S., 2015. FeCl3 pretreatment of three lignocellulosic biomass for ethanol production. ACS Sustainable Chem. Eng. 3, 1794–1800. Conde, T., Mussatto, S.I., 2016. Isolation of polyphenols from spent coffee grounds and silver skin by mild hydrothermal pretreatment. Prep. Biochem. Biotechnol. 46, 406–409. Devi, Y., Thirumdas, R., Sarangapani, C., Deshmukh, R.R., Annapure, U.S., 2017. Influence of cold plasma on fungal growth and aflatoxins production on groundnuts. Food Control 77, 187–191. Fenton, H., 1894. LXXIII.—oxidation of tartaric acid in presence of iron. J. Chem. Soc. Trans. 65, 899–910. Grzegorzewski, F., Ehlbeck, J., Schlüter, O., Kroh, L.W., Rohn, S., 2011. Treating lamb’s lettuce with a cold plasma–influence of atmospheric pressure Ar plasma immanent species on the phenolic profile of Valerianella locusta. LWT Food Sci. Technol. 44, 2285–2289. Haripriya, R., Selvaraj, C., Naveenraj, D., Kirubakaran, S.A., Muthukumar, V., Thirumalaivasan, P., 2014. Pretreatment of cellulosic waste materials. Eur. J. Biotechnol. Biosci. 2, 4–13. International Coffee Organization (ICO), 2016. The Current State of the Global Coffee Trade. Jahan, M.S., Saeed, A., He, Z., Ni, Y., 2011. Jute as raw material for the preparation of microcrystalline cellulose. Cellulose 18, 451–459. Jaiswal, A.K., Abu-Ghannam, N., 2013. Kinetic studies for the preparation of probiotic cabbage juice: impact on phytochemicals and bioactivity. Ind. Crops Prod. 50, 212–218. Jaiswal, A.K., Gupta, S., Abu-Ghannam, N., 2012. Optimisation of lactic acid fermentation of York cabbage for the development of potential probiotic products. Int. J. Food Sci. Technol. 47, 1605–1612.

Jung, Y.S., Lim, W.T., Park, J.Y., Kim, Y.H., 2009. Effect of pH on Fenton and Fentonlike oxidation. Environ. Technol. 30, 183–190. Jung, Y.H., Kim, H.K., Park, H.M., Park, Y.-C., Park, K., Seo, J.-H., Kim, K.H., 2015. Mimicking the Fenton reaction-induced wood decay by fungi for pretreatment of lignocellulose. Bioresour. Technol. 179, 467–472. Kato, D.M., Elía, N., Flythe, M., Lynn, B.C., 2014. Pretreatment of lignocellulosic biomass using Fenton chemistry. Bioresour. Technol. 162, 273–278. Kwon, E.E., Yi, H., Jeon, Y.J., 2013. Sequential co-production of biodiesel and bioethanol with spent coffee grounds. Bioresour. Technol. 136, 475–480. Laux, C.O., Spence, T.G., Kruger, C.H., Zare, R.N., 2003. Optical diagnostics of atmospheric pressure air plasmas. Plasma Sources Sci. Technol. 12, 125. Lu, H., Patil, S., Keener, K.M., Cullen, P.J., Bourke, P., 2014. Bacterial inactivation by high-voltage atmospheric cold plasma: influence of process parameters and effects on cell leakage and DNA. J. Appl. Microbiol. 116, 784–794. Meiners, A., 2010. Optical emission spectroscopy for plasma diagnosticsApplication note. Andor Technologies. Misra, N.N., Pankaj, S.K., Frias, J.M., Keener, K.M., Cullen, P.J., 2015. The effects of nonthermal plasma on chemical quality of strawberries. Postharvest Biol. Technol. 110, 197–202. Moiseev, T., Misra, N.N., Patil, S., Cullen, P.J., Bourke, P., Keener, K.M., Mosnier, J.P., 2014. Post-discharge gas composition of a large-gap DBD in humid air by UV– Vis absorption spectroscopy. Plasma Sour. Sci. Technol. 23, 065033. National Institute of Standards and Technology. 2013. Atomic spectra database 2012. Available from: [cited 9th June 2013]. Nguyen, T., Zavarin, E., barrall, E.M., 1981. Thermal analysis of lignocellulosic materials. J. Macromol. Sci. Part C 20, 1–65. Pearse, R.W.B., Gaydon, A.G., 1976. Identification of Molecular Spectra. Chapman and Hall. Pignatello, J.J., Oliveros, E., MacKay, A., 2006. Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry. Crit. Rev. Env. Sci. Technol. 36, 1–84. Ramazzina, I., Tappi, S., Rocculi, P., Sacchetti, G., Berardinelli, A., Marseglia, A., Rizzi, F., 2016. Effect of cold plasma treatment on the functional properties of freshcut apples. J. Agric. Food Chem. 64 (42), 8010–8018. Ravindran, R., Jaiswal, A.K., 2016a. A comprehensive review on pre-treatment strategy for lignocellulosic food industry waste: challenges and opportunities. Bioresour. Technol. 199, 92–102. Ravindran, R., Jaiswal, A.K., 2016b. Exploitation of food industry waste for highvalue products. Trends Biotechnol. 34, 58–69. Ravindran, R., Jaiswal, S., Abu-Ghannam, N., Jaiswal, A.K., 2017a. Two-step sequential pretreatment for the enhanced enzymatic hydrolysis of coffee spent waste. Bioresour. Technol. 239, 276–284. Ravindran, R., Jaiswal, S., Abu-Ghannam, N., Jaiswal, A.K., 2017b. Evaluation of ultrasound assisted potassium permanganate pre-treatment of spent coffee waste. Bioresour. Technol. 224, 680–687. Ravindran, R., Jaiswal, S., Abu-Ghannam, N., Jaiswal, A.K., 2017c. A comparative analysis of pretreatment strategies on the properties and hydrolysis of Brewers’ spent grain. Bioresour. Technol. http://dx.doi.org/10.1016/j. biortech.2017.06.039. Rickard, P., Laughlin, T., 1980. Detection and assay of xylanolytic enzymes in a Cellulomonas isolate. Biotechnol. Lett. 2, 363–368. Rocha, G.J.M., Gonçalves, A.R., Nakanishi, S.C., Nascimento, V.M., Silva, V.F.N., 2015. Pilot scale steam explosion and diluted sulfuric acid pretreatments: comparative study aiming the sugarcane bagasse saccharification. Ind. Crops Product. 74, 810–816. Sarangapani, C., Devi, Y., Thirundas, R., Annapure, U.S., Deshmukh, R.R., 2015. Effect of low-pressure plasma on physico-chemical properties of parboiled rice. LWT Food Sci. Technol. 63, 452–460. Sarangapani, C., Misra, N.N., Milosavljevic, V., Bourke, P., O’Regan, F., Cullen, P.J., 2016a. Pesticide degradation in water using atmospheric air cold plasma. J. Water Process Eng. 9, 225–232. Sarangapani, C., Thirumdas, R., Devi, Y., Trimukhe, A., Deshmukh, R.R., Annapure, U. S., 2016b. Effect of low-pressure plasma on physico–chemical and functional properties of parboiled rice flour. Food Sci. Technol. 69, 482–489. Schultz-Jensen, N., Kádár, Z., Thomsen, A.B., Bindslev, H., Leipold, F., 2011. Plasmaassisted pretreatment of wheat straw for ethanol production. Appl. Biochem. Biotechnol. 165, 1010–1023. Scully, D.S., Jaiswal, A.K., Abu-Ghannam, N., 2016. An investigation into spent coffee waste as a renewable source of bioactive compounds and industrially important sugars. Bioengineering 3, 33. Sindhu, R., Kuttiraja, M., Binod, P., Janu, K.U., Sukumaran, R.K., Pandey, A., 2011. Dilute acid pretreatment and enzymatic saccharification of sugarcane tops for bioethanol production. Bioresour. Technol. 102, 10915–10921. Souza-Corrêa, J.A., Oliveira, C., Nascimento, V.M., Wolf, L.D., Gómez, E.O., Rocha, G.J. M., Amorim, J., 2014. Atmospheric pressure plasma pretreatment of sugarcane bagasse: the influence of biomass particle size in the ozonation process. Appl. Biochem. Biotechnol. 172, 1663–1672. Tamaki, Y., Mazza, G., 2011. Rapid determination of carbohydrates, ash, and extractives contents of straw using attenuated total reflectance fourier transform mid-infrared spectroscopy. J. Agric. Food Chem. 59, 6346–6352. Van Dam, H., Kieboom, A., Van Bekkum, H., 1986. The conversion of fructose and glucose in acidic media: formation of hydroxymethylfurfural. Starch-Stärke 38, 95–101.